Rand Fishkin of Sparktoro

Founder Coffee episode 042

Rand Fishkin of Sparktoro

I’m Jeroen from Salesflare and this is Founder Coffee.

Every few weeks I have coffee with a different founder. We discuss life, passions, learnings, … in an intimate talk, getting to know the person behind the company.

For this forty-second episode, I talked to Rand Fishkin, formerly Co-Founder and CEO of Moz and now Co-Founder and CEO of Sparktoro, a new solution and data provider for audience intelligence.

Rand started off leading an SEO agency and blogging about the topic and then started Moz 13 years ago as one of the first solutions for SEO research. A few years ago, he left Moz and started Sparktoro, continuing to work on his love for making big data and complex problems easy.

We talk about SaaS vs. services, the innovative business model behind Sparktoro, how to stay authentic in marketing, how Rand organizes marketing in his new company, and why not to release any MVP but take it slowly.

Welcome to Founder Coffee.

get Salesflare

Prefer listening? You can find this episode on:


Jeroen:

Hey guys. Hey SaaS Growth Hacks. We’re live here on Facebook with Rand Fishkin. For those of you who don’t know, Rand has founded Moz in the past. How long ago was that Rand?

Rand:

Yeah, 2003.

Jeroen:

2003! That’s 17 years ago. It’s quite a while. And recently I think about two years ago or so you launched Sparktoro after leaving Moz. When did you launch it exactly? Was that last month?

Rand:

Yeah, April. So not quite two months ago.

Jeroen:

Yeah. I followed some of that, but I haven’t been able to test it myself so far, I must say. How is the launch going so far?

Rand:

It’s a pretty tough time to try and pull off a lunch, but it’s going okay, I would say. Not fantastic, but not terrible.

Jeroen:

Can you tell some of the viewers and listeners may be what Sparktoro is about, so that they can get an idea of that because it’s still a pretty young company?

Rand:

Yeah, absolutely. Sparktoro is an audience intelligence and a market research tool. It’s designed to help marketers, and product people, and entrepreneurs and founders better understand the audiences that they want to reach, and where they can reach them online. If you and I are building a product and we want to target chemical engineers in the UK, or interior designers in California, or project managers in the tech field, or any describable audience, one of the key questions that we always have to answer is, where can we reach them? What are the websites they visit? What are the social accounts that they follow? What are the YouTube channels that they subscribe to? What are the podcasts they listen to?

Jeroen:

Exactly.

Rand:

That’s really hard information to get right now. Most people do it through surveys and interviews and googling around, and it’s pretty poor quality and pretty non statistically sound. And so Sparktoro solves that problem by crawling tens of millions of social accounts. I guess billions of social accounts and data points, and then aggregating all that stuff together into these 75ish million profiles that you can search across and say, show me what interior designers in California listen to, watch, read, follow, subscribe to.

Jeroen:

Yeah, cool. Actually, when I think of it, in my previous job, I used to serve these pharma companies mostly, life sciences companies. And what we would do is we would do these expensive surveys, I must say. In which we would set up a survey towards doctors to ask which channels they get most impacted by. For instance, if it’s magazines, what kind of magazines they read, all those kind of things. We then mapped out all the channels through which a doctor is influenced, and then made sure that the pharma companies put their money in the right places. So that’s really what you do, right?

Rand:

Yeah, exactly. Obviously, surveys and interviews have lots of other purposes, but we don’t think they’re very effective for finding out where you can reach your target audience with your marketing. And if you don’t want to just throw money at Google and Facebook or Amazon, if you’re an eCommerce and sort of pay these incredibly inflated prices, try and compete against thousands of other competitors. If you want to go direct, if you want to do organic marketing, co-marketing, pitches of all kinds, there is tremendous opportunity to have a competitive advantage in the marketing world if you know which sources reach your audience.

Jeroen:

Yeah, when I think of it, I’m certainly going to have a look at it for Salesflare as well to see how we can get to many people.

Rand:

The bad news is that Sparktoro is primarily somewhere between primarily and exclusively English language centric. So if you’re looking at UK, Canada, Australia or United States, some South Africa, New Zealand, it works pretty well. If you’re looking outside those places, not as much.

Jeroen:

Right now our main markets are English spoken, so that sounds good. But what I actually find interesting is that both at Moz and Sparktoro, you’re sort of doing things in a similar way. You’re like indexing traffic drivers of the web, you could say. Earlier from a search standpoint and now more from a social standpoint. What is it exactly that attracts you towards these type of businesses?

Rand:

I like making big data and complex problems easy. I think that’s a valuable thing to do. I have fun doing it. I enjoy being able to aggregate these huge amounts of disparate information, and distill it down to something useful and usable. I think it speaks to me for whatever reason.

Jeroen:

And what is exactly the backstory there for starting Sparktoro? Not like the fluffy version of how you always wanted to do this. But really concretely, when did the first spark for Sparktoro – pun not intended – happen? And how did that start to grow on you?

Rand:

The first spark was or the catalyst was certainly a few years ago. I was on the board of a company called Haiku Deck here in Seattle. We were consistently struggling with marketing channels and tactics. Essentially, on occasion, the CEO and the team would stumble into one marketing channel that worked really well for them, like a group for whatever professional life coaches, which was like a big target audience for them. And they’d have a great month or even a great quarter, and then it would be this constant search for like, “oh, can we find another one?” How do we figure out which, whatever, the conferences and events, YouTube channels, podcasts that were good for them, a lot of websites sponsorship stuff was good for them, co-marketing stuff was good for them?

Rand:

Sometimes a social, the right social account amplifying would be good for them. But figuring those out was just like pulling teeth. And so I remember having a conversation with Adam the CEO and saying like, gosh, I think this data is possible to get if we could just crawl it. And we had this conversation and like, wait, are we really going to build a web crawler and a social media crawler just to be able to extract the information for which channels we can go after? Like, that’ll be a year of work to build this thing. It never got made, but it definitely sparked the idea in my head that it could be done.

Jeroen:

Cool. I’m really fascinated by trying to quantify social influence, and some companies have tried to do it in the past.

Rand:

Yeah. I mean, I think those are sort of solving a different problem, where they’re cloud, and who’s it? Folks like Brandwatch. And people in those spaces, they’re looking at sort of, what are the social media conversations that are happening? Or how many followers does this person have versus that person on these different accounts? And Sparktoro is really, it’s very simple and straightforward. It’s just challenging to get. Essentially, you’re saying ‘oh, does this source have lots of followers’, ‘does it have lots of impact’? You kind of don’t care about that.

Rand:

Like when you’re a brand what you care about is, does it have impact with ‘my’ audience, the people I want to reach? So, you don’t care if it has a million followers. Influencer marketing is so terrible at this because influencer market is all like, ‘oh, let’s find a person with whatever, 100,000 or a million Instagram followers, and we’ll pay them $500 to pose half naked with our product’. And it’s kind of ridiculous. Does that actually move the needle? For some consumer companies sometimes it does fine. But in a lot of cases, if you’re targeting a more specific audience, that’s just a terrible solution. What you want to know is not, who has a million followers. It’s, who do the 10,000 people that I want to reach, follow and pay attention to the most?

Rand:

That’s the real question you want to answer. And so what you’re looking for is, I want to be able to search for my audience in ways like words and phrases in their bio or their profile, things that they talk about online, sources they already follow, hashtags they use, websites they visit. Those types of searches, and then see a list of the people, and publications, and accounts and podcasts that they follow in order. So, 18% of interior designers in Los Angeles follow or listen to this publication.

Jeroen:

Cool.

Rand:

So, now I can do good targeting. I know that I can put three times the budget to this one that I can put to the one that says six percent.

Jeroen:

Yeah. And how does that system exactly work? So, you were saying some of the things, if I remember well, it’s hashtags they use. You also said things that they have in their bio, but you also said sites they visit.

Rand:

So, implied visit. So by crawling social accounts and about pages, we just look at what they link to and engage with, and make the assumption fingers crossed. Hopefully most of the time, if you post something in Facebook, or Twitter, or LinkedIn, or Reddit or YouTube. If you post a link, chances are good you probably visited that. So that’s an implied visit. We don’t actually have a monitoring system installed on anyone’s computer. This is implied data from accounts – public social and web accounts.

Jeroen:

So do you crawl the whole history? Like Twitter trends that every person follows, forever?

Rand:

No, not entirely. We crawl a pretty significant portion of Twitter. And Twitter’s where we get a lot of data. But we basically are crawling a sample of the last 120 days of what’s been shared, engaged with, linked to, followed, talked about on Instagram, Reddit, YouTube, LinkedIn, Twitter or Facebook. Quora, I think is one of our networks or Medium, website about pages. So we’ve got about 10 social networks, and then websites.

Jeroen:

How does the data coming from Twitter compare to the data from many of these others? I imagine that for Twitter, it’s way better than for instance, like crawling Instagram?

Rand:

No, Instagram and Twitter are both pretty good.

Jeroen:

Yeah.

Rand:

Instagram and Twitter are both pretty good. Facebook is challenging because a lot of the information’s private. Reddit is challenging because many folks don’t publicly connect their Reddit account to any other social account. So Reddit, is sort of a very anonymous network. Medium is pretty good when we can get it. If someone has a Medium account, that does work pretty well, but it’s a small portion. YouTube is actually pretty darn good. Because you visit someone’s YouTube page, if it’s public, it’ll show what they’re subscribed to in their comments and that kind of stuff. You can see channels, blah, blah, blah. So every network has its challenges, every network has its opportunities. LinkedIn has nice data about job title and description, all that kind of stuff.

Jeroen:

Yeah. Cool. You must be sitting on an enormous amount of data by now already. Like, you’re probably able to see trends in there that are related to the current situation. Are there are any trends you’re seeing in the data you’re crawling about, for instance, how marketing works?

Rand:

Sure. Yeah. I think with regard to the pandemic, there’s folks who are doing some, I would say more relevant and better stuff on that data extraction front. I’ve been really impressed with HubSpot’s COVID-19 analysis of sort of how email, how effective are B2B emails, what are the open rates, what are sales, lead rates, what are close rates like. I really liked what SimilarWeb is doing in terms of traffic data. Stuff that we see is more like more people following COVID related accounts, following scientists and health organizations. We see people tweeting about things, sharing, publishing data on LinkedIn, publishing on their public social profiles, so there’s more interaction around that stuff.

Rand:

I think there’s more activity in general. But I would say, I actually noticed the last month or so, most of that activity is not pandemic related. It is related to the Black Lives Matter movement. And that’s certainly an area where we can see some really interesting data. Yesterday in fact, I saw an account – this is a New York Police Sergeants’ Union, so an organization. It’s an official account. And I plugged in that account to Sparktoro to take a look. They had amplified a video from a couple of weeks ago saying that it had been from the night before to try and I don’t know, stir up pro police anti black sentiment among their followers. And it was very manipulative, an obvious lie, really problematic, really troubling because this is an official account for a police union.

Rand:

I plugged in their account, and sure enough, if you look at the data about their followers, the people who follow this account, they were white supremacists, believers in, I don’t know if this has made its way to Belgium but do you know what QAnon is?

Jeroen:

QAnon? No, I don’t know.

Rand:

So there’s like a substantive subset of conservatives in the United States who believe that Hillary Clinton is allied with underground lizard people in a pizza dungeon.

Jeroen:

Okay.

Rand:

I know you’re laughing. It’s a real thing. There are a lot of people who believe that there’s this like numerology thing, and Trump is somehow going to save the world from these lizard people, and it’s pretty weird. There’s a lot of pedophilia accusations. It’s very odd, very hard to believe.

Jeroen:

Oh, I saw some of that part. It’s an old Jeffrey Epstein thing with the pedophilia, yeah.

Rand:

Yeah. I think once that came out, that the Jeffrey Epstein things started getting associated with this, the rest of the QAnon movement. But it’s a very broad, strange conspiracy theory about how the world is controlled, and the deep state and all these kinds of things. Anyway, all of those things, there was a bunch of what’s it called? Like words and phrases people use to describe themselves, acronyms they were sharing, hashtags they were using. And I didn’t even know what they were, so I had to Google them to look them up to figure out like, oh, okay, a bunch of these folks think that the Black Lives Matter movement is connected to QAnon and, whoa. So, you can kind of see the underpinnings of an audience.

Rand:

And this is around a political topic, obviously. But what’s very useful for marketers outside of the political spectrum is things like being able to understand the words and phrases that your audience uses. If you don’t understand their acronyms, you don’t know the words that they’re using, you don’t understand how they’re talking about things and what they’re paying attention to, you can be part of that conversation and listen before you go try and amplify your messages. So it is helpful not just for discovering sources, but also for gaining a deeper understanding of your potential buyers.

Jeroen:

Yeah. So if one uses your software in a certain way, they can also discover these kind of messages?

Rand:

Yeah. I mean, basically, we’ve done a number of analysis but one of the most popular things that people did, we have a tool that analyzes fake followers to look at what percentage of a person’s social account is genuine. The free tool is just for Twitter, but what percentage of my followers are not active? Essentially, they’re bot accounts, they’re propaganda accounts, they are non active accounts, they’re spam. And so we have this whole analysis that we run through. And almost everyone who uses the tool does two things; they check their own account, and then they checked Donald Trump’s account. Like same thing every time.

Rand:

And yeah, about 70% of Donald Trump’s followers on Twitter are somewhere between not real and not active accounts. They’re spam, they’re bots, they’re propaganda. Or they’re just not active. They signed up at some point in Twitter in the past, but they haven’t logged in 100 and plus days.

Jeroen:

And all these likes that he gets on every tweet that he sends out, are those real or bots’ likes?

Rand:

A combination. Yeah, a combination.

Jeroen:

I’m always surprised. He says something really surprising, and then he gets like a ton of likes. It’s a special world.

Rand:

Yeah, it’s pretty problematic. I think the good news is Twitter is unlike Facebook, willing to take some steps to limit the visibility of his hate and violence inciting tweets. And being that they’ve said publicly that if he were a private citizen and not the president, they would have banned his account because it’s violating their terms of service by inciting hate speech and inciting violence. And yeah, that’s pretty concerning. It’s behavior we’ve never seen from a president before.

Jeroen:

True. About all this, I see that you’re often tweeting about things like the Black Lives Matter movement, and talking openly about politics. And while a lot of people would advise against doing that in the context of business, how do you exactly look at that?

Rand:

Yeah. So, I have two beliefs. One, I think it is a moral and ethical imperative for a thinking, caring human being to put their beliefs about the world that they want to see and the world that they want to live in above their desire for additional financial gains. I’ve been unimpressed, somewhere between unimpressed and disgusted by folks who are sort of willing to say, well, privately yeah, I support that. And I think police violence is bad, but I’m not going to say anything about it publicly because I don’t want to piss off potential customers. Maybe I have a lot of money and I’m a business owner, but I want more money more than I want to take a stand on an issue. That’s pretty gross. That’s the wrong side of history. That is really disappointing.

Rand:

And so yeah, I am not willing to be in that world. And I don’t have that much interest in money that I’m willing to make that kind of sacrifice. And also as a marketer, one of the things that I think has always been true is that building a community around yourself that supports and believes in what you’re doing for more than just what your business or product provides is a good thing. I think, they have many, many faults. But Nike in the United States supported Colin Kaepernick, the American football players’ protests against police violence and his kneeling during the national anthem. They were willing to take his side and be supportive, and amplify his message in their branding publicly, despite being a company that is bought by many slots of the American public. And potentially, I think, maybe in small ways, maybe even in larger ways, helped to make that conversation more accessible and palatable to a much wider swath of American citizens.

Rand:

I think that Nike is imperfect in many ways. I have many complaints about things that they’ve done in other sectors, but I think that is somewhere where you could point and say, hey, that’s a good thing. Ben and Jerry’s ice cream has taken a shockingly wonderful stand on a whole bunch of these issues. You can see just this past week, NASCAR stepping up and saying, we’re banning the Confederate flag from our event. We’re not going to have this representation of hatred and anti-blackness at our public races anymore. And yeah, I mean that these are things that I think people in leadership roles should contend with, should be forced to contend with, and they have the option to make the right decision. I hope that more people do.

Jeroen:

Yeah, I noticed that you’re a big proponent of more authentic and personal marketing. I hate to call it personal branding, because that just makes it salesy again.

Rand:

One of the things that’s really interesting, I don’t if you’ve experienced this or how much you’ve been talking about this stuff on your public social feed, or even how much it’s made its way to Belgium, but I have lost the most social followers in the last four weeks that I’ve ever had in my career. People stopped following me on Twitter, on Facebook on LinkedIn. And well, I mean, that sucks, but also, I’m probably better off reaching the people that are right types or share the same vision for the world that I do.

Jeroen:

Mm-hmm. Okay. Let’s completely switch topics now. Something extremely different, and it’s about SaaS versus services.

Rand:

Oh, yeah.

Jeroen:

I read through your book, Lost and Founder. I got it here just to show a while back, and I really enjoyed it. Thanks for writing it. And I remember that in there, you had a very interesting view on SaaS versus service/agency businesses. In the group also that we’re broadcasting this, there’s a lot of SaaS businesses but there’s also agencies in there offering services to SaaS businesses. And in both cases, or sort of as a SaaS businesses, people tell us to hold off for services. And as a service business, you might want to go into SaaS business. But then on the other hand, it sort of feels like you should cancel your service business.

Jeroen:

You gave in the book some differences between the two, and why you should choose for the one and the other, and that they’re not mutually exclusive. Perhaps, can you shed some more light on that? Summarize your thoughts on the difference?

Rand:

Sure. If we’re talking about building these two types of businesses, my opinion is that one is not necessarily better than the other, one is not necessarily going to make you richer than the other. And there’s a lot of false belief around this. I think a lot of that is spread unfortunately, from the venture capital tech world. So there’s a lot of information out there that’s like, oh, an agency business is whatever, a nice lifestyle business, or it’s not scalable. Or it is, you’re never going to get the profit margins of software as a service.

Rand:

Then you look at venture backed SaaS businesses, and their technical gross margin is quite high. But net is often barely break even. And often, even more often, they lose money for a very long time until they go public or get sold. I mean, HubSpot was losing money until what, a year and a half, two years after they IPOed? For their entire history, they lost money until then. It is really, really important to think about what you’re passionate about, and what you want and the kind of business that you want to build, and not be biased by tech and media sources and the venture capital world. I would encourage very, very few people to go out and seek Venture Capital.

Rand:

I think that services businesses can sometimes build great software products on the side, and offer those and have them be a part of their competitive advantage. And I think that software businesses can offer services, and have that be part of their competitive advantage. These are not mutually exclusive things. I think anyone who tells you otherwise is kind of 10 years behind the times. There was a period from the early 2000s and mid 2000s, until the mid 2010s when for some reason it was popularized that service is bad, software product good, never should they meet. And now everyone sees that SaaS businesses at scale almost always offer services on top of them, and services can help SaaS business have higher lifetime value, and better retention rates and lower churn.

Jeroen:

Yeah. Are you offering any services next to the SaaS businesses at Sparktoro now, or is this a plan eventually?

Rand:

No, we’re not right now, and it is not part of the plan for us. That being said, we offer, what will I call it? When you subscribe to a package, you’ll get an email, a personal email from me that I just write to everybody. And I work with almost everyone who signs up to help them use the product and help them with their marketing. A lot of them, I’m helping them find agencies and consultancies, a bunch of agencies and consultants. I’ve been helping them apply the data to their client projects. So yes, you get a very personalized level of service from the founders. And that will probably be true until we have a few hundred customers.

Jeroen:

Yeah, that’s for free for now.

Rand:

Right.

Jeroen:

Cool. Why is it actually that you chose to run a SaaS business? Because you used to run an agency way back, like more than 17 years ago, I guess.

Rand:

Yeah, it’s mostly because I don’t like sales. I don’t like the process of sort of selling people directly one to one. Then also, I don’t like the process of keeping them as customers by continuing to sell them and upsell them. That’s just not an interest of mine. And also I really do as we talked about earlier, I really do like building software and aggregating data, making that available to folks. And so yeah, because that’s a passion that made sense as a business model.

Jeroen:

Yeah. In your book, I also remember you made a financial comparison between the two. Anyone thinking about making a choice purely for financial reasons, what would you advise, agency or SaaS business? It’s a hard question, I know.

Rand:

Really a hard question. I mean, the weird part is a SaaS business, if it’s successful and it’s at scale, and you can find a buyer for it, these are all big ifs. Assuming that you own almost all of it, you probably will come out better than an agency business with similar growth numbers, similar sort of scale and size. That being said, it is often easier to sell an agency than it is to sell a SaaS business. It depends. Actually, you know what? That’s not true. It really depends on what you’re in. Some agencies are very easy to sell because they’re in a market that a lot of bigger agencies want to buy in. Some SaaS businesses are in a particular market, and so they’re easier. So, I don’t know that there’s a balance there.

Rand:

I would say, with the two models, the degree that the likelihood that you will be able to successfully build and scale up a business is much lower on the SaaS side than it is on the agency side. Most agencies are able to successfully get off the ground. Consultancies, independent consultants, they’re able to get off the ground. They survive longer. They tend to be profitable for longer. SaaS products, high failure rates, most of them don’t get off the ground. Many that do get off the ground can’t sustainably find customers and fail in their first year or two. So a pretty and very, very hard comparison to make. If you already have a ton of money and you’re trying to like build one or the other, this is why I tell people to go for their passion. Go for what interests you more, you’ll probably be better at that anyway. And it’s really about what you’ll be better at. The comparison of well, I could build a million dollar a year agency or I could build a million dollar a year SaaS. It’s not comparable.

Jeroen:

A bit further on that in a way, I also noticed that even though you’re selling SaaS with Sparktoro, you’re also selling these non subscription packages outside, these one off packages, which are then against the idea of creating recurring revenue. Can you tell us something more about the idea behind that, and how that actually impacts your business model and results, I would say?

Rand:

Yeah, yeah. So we heard from a bunch of our early, like beta customers and people we surveyed and interviewed that one of their big frustrations, especially on the agency side was SaaS products that only offered subscriptions and you couldn’t buy one time use packages. And that’s because a lot of those agencies use a tool specifically for a client project, and wish they could pass billing on to their client directly. Like, hey, this is a cost of us doing this work. A lot of people have to do that with Sparktoro. So we said, hey, let’s offer one off packages to support that use case. We’re not venture backed. We don’t plan on raising venture, so we don’t have to look at our, whatever. We don’t have to care about the venture metrics. We just have to care about, is our business profitable? Is it doing well? Is it serving customers well? Is it making people happy?

Rand:

And we believe that a lot of people who buy a one week package will come back in the future and buy another one week package when they need this again. In fact, we’ve already seen that. We’ve seen a couple of agencies that have done exactly that. They buy two one-week packages over just the last two months.

Jeroen:

Yeah, and you can maybe even charge them more because they charge it to the customer anyway.

Rand:

Right. They’re charging directly to their customers, and so we just give a high level of use and that it’s not recurring. And yeah, we think about lifetime value just really differently than a classic venture backed SaaS business would. So lifetime value for us is over the next five years, how many times will someone subscribe and unsubscribe? How many packages will they buy? How many people will they tell about it and refer? It’s a much more holistic, big picture, broad view of life value versus, just a credit card connected to an account. And how many months were they paying?

Jeroen:

Yeah, which is easier to track but is less complete. Are you thinking about tracking that? Do you have a system in place for that, or you’re also that data?

Rand:

We have a rough system in place. But no, I think to be very frank, a lot of that is just based on faith. Like, we are not worried about having perfect data to be able to prove it because we don’t have to prove it to anyone. We know that it is true that those things will happen, and so we invest that way. We do have ProfitWell, which is really nice tracking for the stripe, credit card side of things, and helps us keep track of ARR, MRR all that.

Jeroen:

You do sort of have investors, right?

Rand:

Yeah, we have 35-36 angel investors. They own units of distribution in the LLC. And basically, we have a very unique model where if the company sells, they get the greater of either their investment amount back or they get their percentage ownership in the company of whatever the sale price is. And we also have a model where we sort of take our first 1.3 million in profit, and we plan to sort of pay those folks back. Then once we do, everybody participates in profit sharing pro rata, founders included.

Jeroen:

Well, so that’s a lot of information to take in.

Rand:

Yeah. My advice for anyone who’s interested in checking on the funding model is Google Sparktoro funding, and we actually open sourced our investment documents. And a few other startups have used those documents to raise money. And I detailed in the post kind of how it works. But the core of it is like, you raise some money from private investors, they own units in your LLC. Once you pay them back, everybody participates in profit sharing. And if the company ever sells, it works just like any other business, everybody makes the amount of money equal to their ownership percentage.

Jeroen:

Okay. So it’s a combination of actually being an owner, plus getting the profit on top of that. So you’re actually getting more than you would normally get by getting shares, not maybe in terms of money, but being able to make decisions?

Rand:

Yeah.

Jeroen:

But it is, and financially, you’re giving more things?

Rand:

Yeah. I think I agree with you. It is more sort of upfront generous to investors. But it’s also quite generous to founders, like Casey and I will get to participate in profit sharing. Any year that we have profits, we can choose whether we want to reinvest those in the business’s growth, or whether we want to pay it out as dividends. The only downside, really the only downside to this funding model is you don’t get the long term capital gains tax rate on the profits that you make from Sparktoro. Look, I think it’s fundamentally unfair and unreasonable that there’s these different taxation rates, it’s just, it’s wrong. But rich people lobby the US government to get the special deal, so that their special investments would get taxed less than everybody else’s.

Rand:

You know what? When you see investors like Warren Buffett complaining that it’s ridiculous that I pay less in tax, a lower tax rate than my secretary, he’s right. It’s fundamentally unfair. It’s unjust. It’s not right. So, Sparktoro does not give you that advantage that venture capital does. But in every other way, I think it’s not only more generous, but also the thing that excites me the most about it is, it does not require billion dollar outcomes or death that bifurcated venture outcome to make money. Point, what is it? Point six percent, point zero six percent, I think of that of 10,000, or 100,000 venture backed companies are going to make, are going to be those unicorns. And most everyone else, I think it’s 95 or 96% of venture invested firms don’t return their minimum capital requirements.

Rand:

So that’s not a good time for any of the employees, it’s not a good time for the founders, it’s not a good time for the customers. It’s not a win for the venture investors. It’s a loss for 96% of the companies and a win for four percent. It sucks. That model sucks. I don’t want to sign up for those odds. I’d much rather have a company that has a long term, sustainable, profitable path to survival and potential growth over the long term, and that can pay out profits when it has profits. And if it turns into a big hit, which Sparktoro could, it’ll still be a huge win for our investors. They just won’t get the, whatever, 12% tax break or 13% tax break, whatever.

Jeroen:

On the final sale, they don’t, or?

Rand:

No. On the profits, they don’t.

Jeroen:

On the profits, yeah. But the profits is what makes it interesting for them to go for another growth or all costs model, towards a profitable company model. Right?

Rand:

Exactly, right. Because every year that Sparktoro is profitable and in business, they’re making some money. Yeah, and potentially. But what we hope, what we very much hope is that the Sparktoro model inspires a lot more people to use this system, and to build companies that have a higher chances of survival, and that don’t have to be unicorns in order to make their investors happy.

Jeroen:

Yeah. How does the governance exactly look around this model? Like, who makes or influences the decision? Do you have a board now based on these angels?

Rand:

Nope. The decisions sit with me. I’d had some frustrating experiences from my past, and this was one of the things that I basically told our investors just like, you’re going to have to faith in me. There’s a few covenants around, like we can’t raise another round of funding unless we get, I can’t remember I. think it’s a majority of our investors, so we’d have to get like 17 of our investors to say ‘oh, yeah that that looks fine’. Gosh, there’s a couple other covenants. They can’t go resell their shares without notifying us. Pretty standard protection stuff.

Jeroen:

Did you exactly select the investors? Was it potential customers and resellers, or was there another criterion?

Rand:

We didn’t have a lot of criteria other than it had to be accredited investors. We did get a lot of folks. I think half or more of our investors are in the web marketing field, which is really nice. Right? These are people who can help us, and help amplify Sparktoro, and yes, who will be customers. A number of our investors are owners of agencies, and their agencies are our customers, so that’s cool. But yeah, we are pretty passionate, I’d say about having this group of people who can help us out. They’re not making decisions, but they are happy to be helpful. And I think that’s what you want in an angel investor anyway. Even if you’re venture backed, oftentimes your first round of funding comes from angels. And those people are not very involved in the business but sort of happy to be helpful.

Jeroen:

Yeah. We have some of those on board. Actually a last question about all this. Why did you go for accredited investors? Because the second part of my question is, do you think you could do this, use this model in a sort of crowdfunding model as well?

Rand:

Yeah, absolutely. In all honesty, we considered doing a crowdfunding campaign for Sparktoro. And ultimately decided against that, but I think absolutely. Probably more possible to do that if the number of dollars that you’re trying to raise is smaller. So, we were looking to raise a little over a million. If you’re looking at 100,000 to 250k, I think crowdfunding is very, very achievable, assuming that you’ve got some credibility with your community, and notoriety in the field and you’re building something that people know that they’re going to want. I think Sparktoro was much more experimental, and we didn’t even know if the technology would work when we started it. And so, we had to take a high risk on that front.

Jeroen:

Okay. Onto another topic then. I have a personal and selfish question for you, that probably is generalizable to a lot of the businesses of our listeners. Imagine you have a CRM company that is way more human easier, more automated than other CRMs. But you’re operating in a very crowded market with a series of large players that have tens, hundreds of millions, billions in some cases, resources available. What would you advise as a marketing strategy to pierce through all this?

Rand:

I think you have to deeply, deeply specialize. You have to be the best CRM for one very particular kind of business, someone who needs something that’s very different from what everybody else offers. And then you basically position yourself that way, build your product features that way, market to those people. Be in all those places, so that could be ‘hey, we serve, whatever accountant’, ‘we’re the best CRM for accountants’, ‘we’re the best CRM for small doctors’ offices, fewer than 10 doctors in an office’, ‘we’re the best CRM for web marketing companies’. So whatever it is. You have to deeply, deeply specialize, that’s what’s going to be the win for you.

Jeroen:

And why do you think this is the case? I’m interested in some more.

Rand:

Yeah, because I think, when you’re small and scrappy, when you don’t have millions and billions of dollars to throw against marketing and branding, to throw against product features and building stuff, it’s unreasonable to ask you to compete in the broad field. And I think that you have to find your way into a space sort of, geographically, or feature wise, or customer archetype wise or positioning wise. You’re not just going to be able to come out and compete with Coke and Pepsi by launching, whatever, Jeroen Cola. It’s not going to happen, right?

Jeroen:

Yeah.

Rand:

But could you launch something locally and get it in a few restaurants with restaurant tours that you know, and start to get it on a couple of grocery store shelves and even independent groceries that you know. And like, take this positioning of oh, well, it’s all whatever, it appeals to organic and ecologically conscious folks, or it appeals to whatever sports fans, or it appeals to energy drink, whatever. Like, you need a differentiator. And I think in the CRM space, the best way to play that is to choose a customer target that is unique, because then you can be in those unique places. And you can serve those people better than a generic one could.

Jeroen:

Yeah. Cool. Well, that’s something we’re thinking about and well, working on actually.

Rand:

A book recommendation for you if you haven’t yet read April Dunford’s Obviously Awesome.

Jeroen:

I have.

Rand:

I highly recommend it.

Jeroen:

We were co-speakers at a conference last year.

Rand:

Oh, yeah?

Jeroen:

When there were still conferences live. Yeah, we could have pasta together and all. Now when you’re co speakers at a conference, you’re both in your home somewhere and it’s a different kind of thing.

Rand:

Very different.

Jeroen:

Yeah. What is it, and what’s actually your marketing strategy with Sparktoro? How many people did you lineup pre-launch, and how did you go about that? Which channels did you use? And how did the funnel work?

Rand:

Yeah, email was huge for us. So we built a very big email list, about 20,000 folks who were on that list, and we built that up over almost 18 months prior to launch. A lot of that came through content, events and conferences, podcasts. To be totally honest, I did a lot of podcasts and discussions, did some big content pieces that brought people to the site, talked about it on my public social channels. Then all of that came back to our email list, and that email list is how we did our early access launch from February to early April. Which, of course, was a terrible, terrible time because the economy was just falling off a cliff and marketers are getting fired left and right, budgets are getting killed.

Rand:

But yeah, so that email list for us was relatively effective despite the pandemic. Then we’ve been relying on that system ever since. So we offer a freemium version of the tool. You can run 10 free searches. When you sign up, we sort of capture you. You can opt in or opt out to having to send you an email, but we follow up with folks who sign up with an email. And yeah, that’s sort of our long term plan, is to continue to nurture people who are continually using and finding value out of the free version of the product, find ways to offer them more value with the paid version and sort of bring them to the higher tiers.

Jeroen:

Yeah. And if you would advise some people who are also looking to launch a product. Basically you said, we got everyone on an email list and you mentioned some of the channels you use. But what were sort of the percentages, how can you attribute the different channels to your success in building that list?

Rand:

I cannot. I really cannot. We don’t have a way to. Podcast is a great example. So you and I are having this conversation, maybe a bunch of folks are listening, and they’re like, oh that Sparktoro thing sounds pretty interesting. I could give a promotional URL, and try and track people through that. I get on stage at a conference or an event, and I give a talk. I go do a webinar, or a training group. I have a copy on a YouTube live event, I’m talking about Sparktoro on social media. A lot of the time, people are not going to click the link or follow the URL that day. It’s going to be a long term. Oh, I heard about it a few times, and then I realized like, ‘oh, shoot! My boss, my team, my client just asked me, hey, should we be doing any sponsorship of websites outside of just Google and Facebook ads’?

Rand:

‘Oh, what was that thing Rand Fishkin was working on? He has a tool like that. Let me search for the… Oh yeah, Sparktoro. Okay, let me go try that’. That’s how a ton of people come to us. And so attribution is very difficult, it’s very imperfect and I don’t care. I am very comfortable with serendipitous marketing, resulting in branding and resulting in people coming to the site. I do not need perfect tracking because again, I don’t need to prove it to anyone.

Jeroen:

No. But I was asking it from a perspective of being able to advise other people.

Rand:

Sure.

Jeroen:

Now they’re all like, oh shit, we have no answer.

Rand:

Yeah. And this is partially true, but not entirely true. Like, I can show you something. If you want, can I share my screen here?

Jeroen:

You can certainly try. Maybe yeah, try it. I hope nothing crashes.

Rand:

So let’s see, it says host disabled, participant screen check. It’s fine. There’s no problem. What I’ll tell you is, I can jump into our Google Analytics and we have set up a free account creation as a goal, a goal in GA. And so what I can see through Google Analytics is last click attribution, so direct sent us, whatever. 266, free account signups in the last week, organic search sent us 243, social sent us 76, referrals sent us 49, emails sent us one. Most of that organic search, in fact, I’m going to guess 80 or 90% of it are branded searches, people searching for Sparktoro or some variant of Sparktoro or something like Rand Fishkin startup, or audience intelligence, Moz founder, whatever. Stuff that basically says, I want to find that thing I heard Rand talking about at some point.

Rand:

Then the direct is exactly that, like a lot of sources that are non attributable. I can go through these and look. I can look at the referral sources, I can see websites that are sending us good traffic. Buffer wrote an article about Sparktoro a month ago, that sends us really nice traffic, like maybe 10 accounts every week. So there’s stuff in there that I can use, but perfect attribution I can’t get. The one thing I would say is if you are trying to convince someone to make an investment, Google Trends can be really helpful for this. So, you can go to Google Trends and you can search for a brand name that might have been mentioned.

Rand:

Let’s say that you had on your podcast last week, I don’t know somebody from Unbounce. And you could go into Google Trends and search for Unbounce and see, ‘oh look, when Jeroen was doing his Facebook Live thing last week, the search volume for Unbounce went up a significant amount. Like he’s driving a lot of traffic. We should get on his podcast. How do we get on this podcast’?

Jeroen:

Isn’t that like, you can see it by the day or by the hour?

Rand:

By the hour.

Rand:

You can see it by the hour, which is really, really nice. So if you see a television commercial during a live sports broadcast, which don’t exist anymore because of COVID. But back when they did, you could see, whatever a football game in the United States and an advertiser running an ad, and you could see the hour when it happened. You could see the spike in their branded search. And so you could intuit how effective was the ad, and how effective was the channel?

Jeroen:

All I’m thinking now is that you could crawl out, like you could crawl the Google Trends thing, get all the data points, and then see when people were mentioned in places and correlate with things.

Rand:

Potentially, yeah.

Jeroen:

It’s quite some work, but yeah.

Rand:

Yeah. I mean, so part of that is saying, hey, can we prove it? And part of it is, I will 100% admit, I cheat. I get to use Sparktoro. Like, I use my own software to see oh, okay, well, where are web marketers paying attention? What are agencies paying attention to? What about market researchers? What about founders of companies? And I look for, here’s the sources that they pay attention to, here’s the podcast, blah, blah, blah. I plug in a podcast or a Twitter account, a website and I can see what their followers look like. Oh, do the attributes of these followers in the audience insights look like the audience that I want to reach? Does that look like people who are likely to be users of our product? It does, but let’s go do that then.

Jeroen:

So, is that how you said yes to this podcast?

Rand:

I mean, I said yes to this podcast, even without checking because I really liked your pitch and your page.

Jeroen:

Thank you.

Rand:

Yeah.

Jeroen:

Can you give us an example of a way in which you use Sparktoro for yourself? Like to inform your marketing strategy?

Rand:

Yeah, one of the recent fields that we wanted to break into was that there’s this world called market research. So that phrase tends to mean people who survey, build large scale surveys, like we were talking about at the start of this episode, and publish them out, and then try and get hundreds or thousands of responses and analyze those. And market research, it is a messy world as you know, but it’s a world that I want to break into. Like, I want people in market research world to be aware of what Sparktoro does and to be aware of us. And so, I have basically used Sparktoro to look for people who frequently talk about market research, who have market research in their profile and bio. Found some influential sources, took those sources and plugged them in as searches and built myself a list.

Rand:

You just check things off in Sparktoro and then say, add to list. And I started following those people on Twitter, I connected with some of them on LinkedIn. Had some conversations, one-to-one conversations. And then what do you know? Got invited to a couple podcasts, did a YouTube channel thing with someone and one of the connections that I made actually via Twitter ended up writing on us. They were writing a blog post about some other stuff, and they included some Sparktoro data in there and link to us. Awesome. Like I’m not even spending any money. I’m not doing any advertising. Most of Sparktoro’s customers are spending ad money, but we’re trying to be very conservative on the ad spending front right now. And so, I’m doing this just organically. And yeah, relatively effective.

Jeroen:

About being conservative, I noticed that you took around two years from funding to launch.

Rand:

Of course, yeah.

Jeroen:

I remember in your book, it’s like chapter one, the second page. What does it say again? Something about great products are rarely minimally viable. Is that the strategy that was behind this?

Rand:

Absolutely. I mean, we knew that the day we launched, thousands of people would look at Sparktoro. And for years afterward, that would be the impression that they’d have about us. That we had that one chance. That those first 30 seconds, 90 seconds that they played with the product to make an impression on what they would take away for the next five or even 10 years.

Jeroen:

Yeah, they would say like, ah, useless.

Rand:

Exactly. They would say, useless. Or like oh, I see what they’re doing here. This is really interesting. Maybe when I need this thing in the future, I will come back to Sparktoro. And so we wanted to make sure that the product was very polished and high quality. That it was easy to use, that the data was solid. I don’t think we did a perfect job, but I think we definitely did a much better job than most of my software launches in the past. I’ve been very guilty in my previous life at Moz of launching some pretty minimally viable MVPs, and Sparktoro is not that.

Jeroen:

Yeah. So not the ship early philosophy. And if you don’t ship early then the question is, how do you still validate your product as early as possible without having to keep it all to yourself for a long while, because that’s really what the whole Lean Startup movement is about, right?

Rand:

Yeah. So instead of going public with it, we had a private launch. All those people who were on the email list, we asked them to fill out a survey. We got information, and then we invited cohorts of them to be beta testers. So private beta testing, and then early access testing behind the scenes rather than public launch. And so people who were in those groups knew that they were getting access to an early version of the product, that they were helping facilitate it getting better and better. That is where we did a lot of our refinement and making sure that we weren’t just building in a closet.

Jeroen:

So these people had more patience with your product.

Rand:

Yeah.

Jeroen:

Good thinking. Good thinking.

Rand:

And Jeroen, I apologize. I’ve only got a couple of more minutes, and then I need to run.

Jeroen:

Let’s start wrapping up then towards learnings. I heard you saying that you read about 50 articles a day, and learn from that. How do you exactly go about that? I cannot imagine myself reading 50 articles a day, especially next to being on podcasts and YouTube, lives and all these kind of things.

Rand:

Yeah. I think that might be an exaggeration on my part. Maybe 50 articles a week sounds probable to me. More like five a day, five to seven a day. And yeah, a lot of that is just finding things that through my social feeds, through Sparktoro has this trending section that I use for a lot of web marketing stuff. That’s really handy, and that’s free. So yeah, that helps me stay on top of what’s going on in the web marketing world. And Hacker News is another big source for me. I read a ton of that. I’m a big Reddit user for information. So, those are my sources.

Jeroen:

They’re socials.

Rand:

Yeah.

Jeroen:

Last question then. I have your book here. But if you would have to point that the book that shaped you as a person the most, or the way you run Sparktoro the most, which book would you recommend to people?

Rand:

Ooh, I mean, we already talked about Obviously Awesome, but I also love yeah, this one, Liz Fosslien and Mollie West Duffy’s No Hard Feelings.

Jeroen:

No Hard Feelings, okay.

Rand:

No Hard Feelings. Excellent. Excellent book, probably the best book on management and working with people that I’ve ever read.

Jeroen:

What is exactly the gist of the book?

Rand:

It’s about embracing emotions at work. Building a better working relationship with the people around you, being willing to have hard conversations. It’s about that emotional sort of soft skills of business, and I think that, that’s under invested in by a lot of folks in our field.

Jeroen:

Yeah. Cool. Thank you. Thank you for sharing all this.

Rand:

Yeah, my pleasure. Thank you so much for having me, man. I really appreciate it.

Jeroen:

Great to have you on Founder Coffee, and to go live together for this Facebook group. It was great.

Rand:

Awesome. All right. Take care of yourself, man.

Jeroen:

Have a good day, y’all.


Enjoyed it? Read Founder Coffee interviews with other founders.

get Salesflare

We hope you liked this episode. If you did, review us on iTunes!

👉 You can follow @salesflare on Twitter, Facebook and LinkedIn.

Jeroen Corthout